1
Real-Time Patient Safety Monitoring in Clinical
Trials
Session #151, February 13, 2019.
Claire Williams -Head of Pharmacovigilance and Amira Thorn -Chief Operating Officer.
NorthWest EHealth.
2
Claire Williams, RGN, MSc
Salary: Currently-NWEH, during Salford Lung Study-GSK
Amira Thorn, BSc, MSc
Salary: NWEH
Conflict of Interest
3
NWEH – Our Story
• NWEH formed 2008 as a not-for-profit collaboration between the
University of Manchester, in collaboration with local hospital and
General Practitioners (GPs-family doctors)
• Medium sized growing company
• Specialising in real world data driven clinical trials
4
Learning Objectives
1. Justify the need for a different type of clinical trial data.
2. Describe what made the Salford Lung Study so pioneering in its research
methodology.
3. Analyze patient safety monitoring challenges and how to use Real World
Studies to assess the effectiveness of new drug therapies.
4. Explain the bespoke IT infrastructure used to support the safety
monitoring and reporting during the Salford Lung Study.
5. Appraise the impact of ‘near to’ real time safety monitoring on the future of
clinical trials.
5
Why is there a need for a
different type of clinical trial?
7
Standard Randomised
Controlled Trials
• Designed with high internal validity
• Conducted in an artificial environment
• Strict adherence
• Restrictive inclusion and exclusion criteria
• Randomisation
Perceived as the “Gold Standard” for generating safety and
efficacy data
8
0
20
40
60
80
100
120
140
RCTs RWE
RCTs
• Bateman, E.D. et al. (2004). Am J Respir Crit Care
Med, Vol 170, 836–844.
• Busse, W. et al. (2008). J Allergy Clin Immunology,
Vol 121(6), 1407–1414.
• Vestbo, J. et al. (2009). Thorax, Vol 64, 939-943.
• Papi, A. et al. (2007). Eur Respir J, Vol 29, 682–689.
Real World Evidence (RWE) Studies
• Breekveldt-Postma N.S. et al. (2008).
Pharmacoepidemiology and Drug Safety, Vol 17, 411–
422.
• Janson, C. et al. (2005). Eur Respir J, Vol 26, 1047–
1055.
• Adams, R.J. et al. (2002). J Allergy Clin immunology,
Vol 110(1), 58-64.
• Stallberg B. et al. (2003). Respiratory Medicine, Vol
97, 835–843.
• de Marco R. et al. (2005). Int Arch Allergy Immunol,
Vol 138, 225–234.
• Corrigan, C. (2011). Primary Care Respiratory
Journal, Vol 20(1), 13-14.
Adherence to Medications in
Respiratory Studies
Greater than
100%!
9
Recruiting a ‘Typical’ Population
Woodcock, A. et al. (2018).
European Respir J, Vol 51.
pii:1701531
10
The
Wild
The Zoo
11
What made the Salford Lung Study so
pioneering in its research methodology?
12
Why Salford?
• Greater Manchester has become one large integrated healthcare
community
• All patients have a unique NHS number
• Most drugs are electronically prescribed
• All inclusive – Salford represents a typical population
• GPs have used electronic medical records since the 1990s
• Sharing patient data between hospitals and GPs
• Enabling linkage of patient level data between healthcare
systems
13
The first study of its kind
initiated with a pre-licence medicine
Maintains the scientific rigour of an RCT
but keeps it as near to everyday clinical practice setting as possible
Uses an integrated Health Record
that facilitates data collection without direct patient contact
An innovative collaboration
between NWEH, GSK, academia, and healthcare professionals
New, J.P. et al. (2014). Thorax; Vol 69,1152–4.
Bakerly, N. et al. (2015). Respir Res, Vol 16,101.
What made the Salford Lung Study Pioneering?
14
• First prospective pragmatic randomised controlled
trial of a pre-licence medicine
• Minimal exclusion criteria
• Bespoke IT infrastructure to monitor subject
electronic health records remotely
• Close to real-time safety reporting
• Endpoints collected directly from electronic
healthcare records
• Minimal intervention outside of normal care
Salford Lung Study
15
Bakerly, N. et al. (2015). Respir Res, Vol 16, 101
2,802 patients
• Patients in primary care
• Aged 40+ years
• GP diagnosis of COPD
• Taking ICS and/or LABA
and/or LAMA
• Exacerbation in last 3 years
• Consented
Study Treatment
Usual care
Randomisation visit
• Routine respiratory review
• Device instruction
• CAT
End of study visit
• Routine respiratory
review
• CAT
12 months
of usual care
Constant real-time data collection of all interventions and
safety monitoring
Randomisation
Aim: To compare the effectiveness and
safety of study Rx to usual care on
meaningful patient-centred outcomes in
COPD (rate of moderate/severe
exacerbations)
During the 12 month treatment period, patients can have their maintenance treatment adjusted (stepping-up,
stepping-down or switch) in both arms at the GP’s/Investigator’s discretion as would be normal clinical practice.
Salford Lung Study Design
16
To maintain the scientific rigour one associates with a
traditional randomised controlled trial
• Interventional
• Randomised
• Control arm
But to keep it as near to everyday clinical practice as
possible
• Embracing heterogeneity of the patient population
• Maintaining patient experience as close to normal as possible
• Collecting endpoints relevant to patients and healthcare decision makers
• Comparing the study drug with ‘usual care’
What was the Intent Behind the Study
Design?
17
• Intensive training of all study and site staff
• Minimal study procedures and visits
• Study drug accessed through community pharmacies i.e.
CVS
How did we ensure we did not interfere
with “normal” care of patients during
the study?
18
• Direct extraction of study endpoints from the electronic
medical record wherever possible
• Excellent and auditable IT systems with available support
staff
• Fully validated system - reduced need for source data
verification
How did we ensure robust collection of
study end points?
19
• Integrated electronic patient record with real-time access
ensures that the safety team are aware wherever and
whenever patient accesses healthcare
• Dedicated safety team
How did we ensure robust safety
monitoring without carrying out routine
study visits?
20
2802 COPD
Subjects
Recruited
130
Community
Pharmacies
Bespoke IT
Infrastructure
3000 Staff
Trained in
GCP
80 GP
Practices
Over 200
Staff Directly
Employed
Specialist
Safety Team
Scale of the
Project
21
235 Million
Rows of Data
73292
Radiology
Results
3.1 Million
Clinical
Observations
15 Data
Feeds per
Patient
3.4 Million
Biochemistry
and
Haematology
Results
1434 SAE
Reports
30200 Safety
Events
Alerted per
Year
6.5 Million
Medications
Processed
Salford Lung Study: Data Collected from Patient Electronic
Healthcare Records (Asthma and COPD)
22
The challenges of monitoring
patient safety
23
Narrow
specific patient
population
Frequent
reviews/hospit
al visits
Adverse
events
captured
through patient
recall
Adherence
encouraged
Monitoring and Recording Safety in
Standard Trials
24
Monitoring and Recording Safety in the
Salford Lung Study
Broad patient
population
Few visits Usual care
Unsupervised
adherence
25
Patient
with
complex
co-
morbidities
High
volumes of
safety data
expected
Minimal
interventio
n beyond
usual care
Key Safety Challenges
Development
of bespoke IT
infrastructure
26
The technology used to support
the safety monitoring and
reporting
27
ConneXon
GP
GP
GP
Hospitals
PAS & EHRs
eCRF
Pharmacy
Pharmacy
Pharmacy
Emergency GP
Study Data
Management
Safety
Alerts
Performance
Management
Regulatory
ONS Mortality
Data
Open Exeter
NHAIS
SUS National
Data
28
Study Nurse tags patient in the Electronic Medical Record
Patient admitted to hospital
Alert automatically sent to the Safety Team - near to real time
Safety Team reviews event & completes Serious Adverse Event (SAE) form
Principal Investigator (PI)
reviews & records causality &
severity & locks SAE
SAE submitted
to Sponsor
Safety Alerting and Reporting
29
Activity Summary - A&E attendance
30
Activity Summary - A&E attendance alert detail
31
Activity Summary-Emergency Admission
32
Activity Summary - Emergency Admission alert detail
33
Activity Summary-Admission event detail
34
Activity Summary- Medical history event detail
35
Activity Summary- Discharge event detail
36
The impact of ‘close to’ real
time safety monitoring on the
future of clinical trials
37
Benefits
• More accurate and comprehensive safety
data
• Safety events processed close to real time
• Configurable to identify adverse events of
special interest
• Reduces the resource required for
investigators and research teams
• Collection of long-term follow-up post trial
Using Electronic Healthcare Records
38
The future
• This platform enables studies to be conducted
in a broader more representative patient
population
• Patients can be closely monitored without
intervention which helps maintain the ‘real
world’ methodology
• Safety events being reported close to real-time,
enabling safety signals to be detected more
quickly than in standard clinical trials
